The Library: Messy by Tim Harford

Reviewed by Matt.

We’ve just finished reading the latest book from Undercover Economist author, Tim Harford.

Messy takes a look at how people and businesses can become more creative and resilient in a tidy-minded world.

The book is an enlightening read that provides some wonderful insights into the surprising benefits of disorganisation, collaboration and improvisation.

Here are a few key takeouts for the marketing and advertising industries.


In an industry that has become more and more obsessed with a data-driven approach, it’s vital that we recognise that using the wrong targets or measurement tools can ultimately lead to poor results.

Harford believes that one of the biggest problems we have with our targets and measurements is that we too often choose the easiest option; the low hanging fruit.

“It’s easier to count trees in a forest than it is to measure biodiversity.”

In advertising we make the same mistakes. We’ve chosen Likes, clicks and (partial) views as indicators for effectiveness.

And when we set the wrong goals we encourage the wrong outcomes. Or worse, when some even look for ways to game the system.

“Consider a bus driver who is rewarded for sticking to the bus timetable. And who keeps on time by driving past lines of passengers without stopping.”

Sound familiar?

These, Harford says, are all cases of tunnel vision that result from targets being too narrow.

And the wrong short term target can have a major impact on the long term goals.

“In each case we assume that by measuring one thing we’re really measuring everything. That is delusional. We hit the target, but miss the point”

Short term targets appeal to us because they feel tidy and the lure of instant gratification is simply too enticing. But an evidence-based approach to marketing shows that the effects of advertising are spread over time. This requires that we take a longer term approach.

Harford’s alternative to a tidy-minded target is;

“One possible approach is to make the targets more sophisticated. Covering more measures, with more attention to detail.”

In marketing, short term sales results are influenced by many factors such as sales promotions and category influences, but when we allow ourselves the time to measure longer term results we’ll get a much more accurate picture of how things are really working for the brand.


In a time of programmatic buying and personalised advertising, Harford warns us against relying too heavily upon automated systems.

“In a messy world, mistakes are inevitable. Yet automatic systems want to be tidy. Once an algorithm or database has placed you into a particular category, the black and white definitions of the data discourage argument and uncertainty. You are a shoplifter. You were parked at a bus stop. You are on the no fly list. The computer says so.”

The outcome of this automated stereotyping not only gives brands a narrow view of their buyers, but it also starts to hinder people’s experiences.

“As a consumer do you want to live in a tightly controlled world where you are served the same classified content? A world where the discovery of new things is not possible.”

It’s very possible that personalisation could lead to a dreary, repetitive world for people.

Retargeting is another example of this type of automation. Banner ads already follow us around the web, despite the fact that we may have already made a purchase or never intend to buy the product in our lifetime. 

“For all the power and usefulness of such data perhaps we have not yet acknowledged how imperfectly a database maps onto a messy world.”

Harford notes that the other major problem with this approach is automation bias.

“Once a computer has made a recommendation, it is all too easy to accept that recommendation unthinkingly.
We worry that the robots are taking our jobs, but just as common a problem is that the robots are also taking our judgement.”

Harford believes that we must look for ways to combine the adaptability and judgement of humans with the reliability of computers. It’s an approach we should certainly give more consideration.


“If we are understanding that mess makes natural systems more healthy and resilient, than could the same be true for artificial systems such as the neighbourhoods, cities and countries where we live?

Hartford notes that diverse industries are better for the economic health of a city. Complex or diversified economies are far more stable than their homogeneous competitors.

“Diverse economies, like diverse German forests are more resilient.”

The same is relevant for an agency. With many agencies moving towards specialised offerings, they’ve lost many of the benefits of having a diverse workplace. Sharing a diverse range of skill sets and experience can help to create unexpected opportunities and discoveries. 

Perhaps if agencies were a little more diverse they would also be more resilient to the unknown changes that the future will inevitably bring. This however, requires a level of openness and improvisation. Unfortunately, as Harford notes, it is a rarity: 

“It is not only children who find themselves nudged and controlled as they wander curiously through life. A good job, a good building, even a good relationship has openness and adaptability. But many jobs, buildings and relationships do not. They are monotonous and controlling. They sacrifice messy possibilities for tidy predictability and too often we let that happen because we feel safer that way. That is a shame.”

It is our desire for predictability and control that ultimately limits us from reaching our full potential. It’s an important lesson that many businesses would do well to learn from.